Habitat Suitability Modelling of wild Goat (capra aegagrus) In Ghamishloo National Park Using Artificial Neural Networks

A species distribution model evaluates the relationship between distribution of species and environmental factors and therefore gives a better understanding of future management plans. Artificial neural network is one of these modeling tools which in these days have become a very important gadget for decision making. According to the International Union for Conservation of Nature (IUCN), Capra aegagrus is one of the venerable species in Iran and it categorizes in VI Section. Therefore, the purpose of this study is to model Capra aegagrus Habitat suitability in Ghomishloo national park using artificial neural network. Different variables which used in this study are altitude,slope, vegetation, distance from the main roads and water resources. In order to analyze the data, multilayer perceptron network have been used and MATLAB R2015b was used for the calculations. The results show that the most important variables which effects on Capra aegagrus distribution are altitude, species distance from water resources, slope, the distance from the main roads and vegetation, respectively. And the results also indicate that the most suitable habitat for Capra aegagrus in Ghamishlo national park is in the areas with an altitude of more than 2200 meters and an area of 3982 hectares which includes 12.36 percent of this national park.